Handwriting is traditionally performed on a writing surface, such as paper, with an ink-dispensing pen or other writing instrument, such as a pencil, or paint brush. The result is expected to be understandable by human readers.
Recently, electronic handwriting has been done on planar X-Y digitizing pads using a stylus employed to simulate handwriting upon the pad to create an electronic facsimile of handwriting. The digitizing system collects an array of X-Y coordinates of pixels corresponding to the curve tracing positional points of the stylus tip. Usually the X-Y arrays are gathered and stored as positional arrays, and are made discernible to a human reader when rendered on an X-Y display, but are rarely discernible as text by a device.
Attempts to make handwriting discernible as machine-readable text have concentrated on handwriting recognition of the X-Y traces by translation into binary coded text after affine transformation of the X-Y trace. Other techniques of recognition of the X-Y traces employ stochastic recognition based on various randomness assumptions using a statistical model. Other attempts with more deterministic techniques of recognition of the X-Y traces use velocity profiling in on-line recognition and forward search in batch recognition. Many similar X-Y trace recognition efforts have resulted in numerically intense algorithms which tend to restrict the recognition process to off-line batch processing, conducted as a separate procedure long after the writing has been done.
More recently, on-line recognition systems have dispensed with natural hand-writing and created specialized pen-stroke shorthand for letters of the Latin alphabet and Arabic numerals and punctuation marks, such as an electronic stylus recognition system. Field experience has shown that recognition error rates are high enough to cause manufacturers to begin supplanting the system with keypads and software keyboards. Miniaturized keypads are slow when compared to normal handwriting speed. Full-sized keyboards, although faster in use than miniature keyboards, are too cumbersome for optimum purposes.
Devices that track X-Y motion in true geometry exist in the form of analog joysticks. These are used as actuators for simulation and as gaming input devices, where a hand-held game controller may incorporate an analog joystick that permits tracking of directional inputs over 360 degrees around an action reference point, and is small enough to be manipulated by a finger tip. The cited range of 360 degrees signifies that the joystick spans a projection of the X-Y plane, but does not span a radial distance, i.e., the joystick is not operable to span a projection along the Z-axis. This is because the range of each joystick sensor is less than the radial range needed to be spanned.
The cited joystick may utilize optical quadrature sensor wheels over two orthogonal axes of rotation. Such a configuration may suffice for directional control over a planar range, but is inadequate for the capture of natural handwriting strokes because the latter requires a depth sensor.
Further limitations and disadvantages of conventional and traditional approaches will become apparent to one of skill in the art through comparison of such systems with embodiments presented in the remainder of the present application with reference to the drawings.